In the present study, a numerically based methodology for determining the load bearing capacity of particulate reinforced metal matrix composites (PRMMCs) under both monotonic and cyclic loadings is presented. A multi scale approach which combines the shakedown analysis with homogenization was used to study the material. To take into account the randomness of the material, statistical methods have been applied to interpret the results. To prepare a sufficient number of representative volume element (RVE) models for the statistical study, a computational tool was developed to automate the generation of RVE samples.
The general work flow of the numerical approach can be summarized as follows: first, a large number of RVE was constructed as finite element models from either real or artificial material microstructures using the aforementioned in-house code. Next, limit and shakedown analyses were carried out by means of the interior-point method. Finally, results were converted to their corresponding macro quantities and evaluated statistically. With this approach a representative PRMMC material, WC/Co, was studied.
Based on the established numerical work flow, ultimate strength and endurance limit of the material were predicted. The relationship between them and other material parameters was examined. The study investigated how the predicted strength is influenced by the RVE size and the size of reinforcement particles. The study also exposed the change of the feasible load domain, when the kinematic hardening of the binder phase is considered or when multiple independently varied loads are applied simultaneously. In addition to that, the study built predictive models to explain what are the decisive factors that determine the endurance limit of the material.